Research On Virtual Machine Migration Based On Multi-Objective Optimization In Cloud Computing Environment

As a new computing model with low price and high efficiency,cloud computing is gradually emerging in various fields.Because of its on-demand service,high reliability,high cost performance,and scalability,it is gradually recognized by individuals and society.In recent years,the scale of cloud data centers has expanded rapidly,resulting in increased energy consumption and costs.Data centers can improve the quality of cloud computing services and save costs through virtual machine migration.Virtual machine migration is a process of redeploying virtual machines based on changes in the physical host load that occur during the operation of the data center.The core of virtual machine migration is virtual machine placement,so it is urgent to adopt a feasible virtual machine placement algorithm to optimize the virtual machine migration strategy.This thesis uses load forecasting and intelligent optimization algorithms to solve the problem of virtual machine placement and implements virtual machine migration accordingly.The main tasks are as follows:(1)A dynamic load forecasting algorithm based on exponential smoothing is proposed,which improves prediction by dynamically selecting smoothing coefficient accuracy.Load prediction for virtual machine placement can better help to select the appropriate destination host.Experimental results show that the load prediction algorithm proposed in this thesis improves the accuracy of prediction,which can reduce the number of virtual machine migrations to a certain extent and decrease energy consumption.(2)In order to solve the problem that the traditional multi-objective optimization virtual machine placement algorithm is difficult to find the optimal solution,this thesis proposes a multi-objective optimized virtual machine placement model for energy consumption,resource utilization,and load balancing.By improving the ant colony algorithm,a virtual machine placement algorithm based on multi-objective optimization is realized.The algorithm can obtain the approximate optimal solution of the model in a limited time,which meets the requirements of the cloud environment for high efficiency and low energy consumption.(3)This thesis applies load forecasting to the multi-objective optimization virtual machine placement algorithm to find the approximate optimal solution more accurately,and then a virtual machine migration module based on the Cloud Sim platform is implemented by combining the algorithm with host load determination and virtual machine selection.